Post on 16-Mar-2018
CONTRIBUTING TOWARDS POLIO ERADICATION IN ETHIOPIA
PAPER I
Newborn Tracking for Polio birth dose vaccination in Pastoralist and
Semi-pastoralist CORE Group Polio Project Implementation Districts
(Woredas) in Ethiopia
CCRDA/CORE Group Ethiopia
Addis Ababa
June 2012
ii
Contents
ACRONYMS .............................................................................................................................................. iv
EXECUTIVE SUMMARY .......................................................................................................................... v
INTRODUCTION ...................................................................................................................................... 1
Health status of Ethiopian under five children ................................................................................. 1
Childhood immunization in Ethiopia ................................................................................................. 1
Polio Eradication in Ethiopia ................................................................................................................ 1
CORE Group Polio Project .................................................................................................................... 2
RATIONALE .............................................................................................................................................. 4
OBJECTIVES ............................................................................................................................................... 6
General Objective ................................................................................................................................... 6
Specific objectives ................................................................................................................................... 6
METHODS .................................................................................................................................................. 6
Study design ........................................................................................................................................... 6
Study area................................................................................................................................................ 6
Study populations .................................................................................................................................. 7
Women with recent deliveries .............................................................................................................. 7
Sample size determination .................................................................................................................... 7
Sampling Procedures ............................................................................................................................ 8
Data Collection ..................................................................................................................................... 9
Health Extension Workers (HEWs) ...................................................................................................... 11
Community Volunteer Surveillance Focal Persons (CVSFPs) .............................................................. 11
Woreda and Health Center EPI Coordinators ..................................................................................... 11
Traditional Birth Attendants (TBAs) .................................................................................................... 11
Community and Religious Leaders ...................................................................................................... 11
Data Entry and Analysis ...................................................................................................................... 12
Ethical Considerations ......................................................................................................................... 12
RESULTS ................................................................................................................................................... 13
Socio-demographic characteristics women who delivered in the previous one year ............ 13
Pregnancy and child birth ............................................................................................................... 16
Maternal Health Service Utilization .............................................................................................. 17
iii
Knowledge of vaccine preventable diseases and vaccination status of children .................... 25
Reasons for not having polio birth dose vaccination .................................................................. 28
Knowledge on prevention, transmission and vaccination against polio ................................. 29
Findings of interviews with HEWs, CSFP, TBAs and EPI coordinators ..................................... 29
Findings from FGD with community leaders .................................................................................. 31
DISCUSSION ............................................................................................................................................ 32
Strengths and limitations of the study .......................................................................................... 38
CONCLUSION ......................................................................................................................................... 39
RECOMMENDATIONS .......................................................................................................................... 40
REFERENCES ........................................................................................................................................... 42
iv
ACRONYMS AFP Acute Flaccid Paralysis
CCRDA Consortium of Christian Relief and Development Associations
CGPP CORE Group Polio Project
CORE Group Child Survival Collaborations and Resources Group
CV Community Volunteers
CHW Community Health Workers
CVSFP Community Volunteers Surveillance Focal Persons
DPT Diphtheria Poliomyelitis Tetanus
EDHS Ethiopian Demographic and Health Survey
EPI Expanded Program on Immunization
FGD Focus Group Discussion
IDSR Integrated Disease Surveillance and Response
IEC Information, Education & Communication
IIP Immunization in Practice
IRC International Red Cross
MLM Midlevel Managers Training
NID National Immunization Days
OPV Oral Polio Vaccine
PEI Polio Eradication Initiative
PPS Probability Proportionate to Size
PVO Private Voluntary Organization
SNIDs Sub-national Immunization Days
SIAs Supplementary Immunization Activities
SNNPR Southern Nations Nationalities and Peoples Region
WPV Wild Polio Virus
v
EXECUTIVE SUMMARY
The CORE Group Polio Project (CGPP) was formed in 1999 and has been active participant in the global
Polio Eradication Initiative (PEI). It has been working in high risk areas of Angola, Bangladesh, Ethiopia,
India, Nepal and Uganda. Bangladesh, Uganda and Nepal have “graduated”. Currently the project is functioning in Angola, Ethiopia and India with fund made available by USAID. The current CORE Group
Polio Project (CGPP) which extends from October, 2007 – September 2012 has a goal of contributing to
polio eradication by increasing population immunity and enhancing the sensitivity of surveillance for
AFP. CORE Group Ethiopia started to function in November 2001 and has been supporting and
coordinating efforts of PVOs/NGOs involved in polio eradication activities.
Estimates of immunization coverage rates in Ethiopia varied widely, but were consistent in that polio
birth dose (Polio 0) coverage is much lower than other antigens. A major reason for the low OPV 0
coverage is the low coverage of institutional delivery in the country in general, estimated at about 10% ,
and in pastoralist and semi-pastoralist areas, in particular. This low coverage deprives newborns the best
opportunity of getting the newborn Polio 0 dose. Other factors or chain of factors that contribute to low
polio 0 coverage have not been studied in Ethiopia.
This research is one of the three studies that have been identified as essential research areas to fill gaps
towards efforts of GORE Group Ethiopia in contributing to polio eradication in Ethiopia.
The main objective of this study was to examine pregnancy and child delivery practices and identify
mechanisms for improving polio birth dose coverage in CGPP implementation districts/ woredas.
The study used quantitative and qualitative methods and included community based cross sectional
study design involving interviews of women who delivered during the previous one year, key informant
interviews with Health Extension Workers (HEWs), program coordinators, Community Volunteer
Surveillance Focal Persons (CVSFPs) and Traditional Birth Attendants (TBAs), and focus group
discussions (FGDs) with community elders and religious leaders.
The study was conducted in 9 districts (woredas) selected using criteria that included representativeness
and feasibility. A team consisting of enumerators, supervisors, coordinators core research team members
and community guides was involved in data collection and ensuring quality after appropriate training
and pretest was done. Quantitative data were entered and analyzed using SPSS version 17. Descriptive
analysis included data presentation using tables, graphs and appropriate summary figures. Appropriate
statistical tests ( Chi squred test) and measures (OR, 95%CI) were used to asses significance and
strength of associations respectively. Multiple regression analysis was used to measure the effect of
different factors adjusted for possible confounders.
The records from the FGDs were transcribed in the language of the interview and then translated into
English for analysis. Data analysis was done using thematic approach on the” Open Code” software
program.
vi
A total of 600 of women who delivered in the previous one year were included in the study. The mean
age of the respondents was 26.3+ SD5.7, median 25 and range 15-49 years. Four hundred ninety (81.7%)
women could not read or write and 56 (9.3%) responded that they can read and write with difficulty. The
great majority of the women (98.2%) were currently married and about 79% had monogamous marriage.
85 (14.2%) were said to have been registered between birth and 14 days.
Three hundred twelve women (52.0%) had attended antenatal care at least once during the last
pregnancy. Five hundred forty nine women (91.5%) delivered their last baby at home and the reasons for
home delivery include familiar birth attendants (46.4%), distance to health facility (24.4%), unavailability
of transport (16.0%), delivery in health facility not culturally encouraged (10.7%) and non-friendly health
services (8.9%).
When adjusted for other factors, religion, having live stock, and income generating activities were
significantly associated with ANC attendance at P<0.05. After adjusting for other factors, religion and no
other income generating activities were statistically significantly associated with home delivery when
ANC was not included in the model. When ANC is included, none of the variables retained statistical
significant.
Polio 0 coverage was 29.7%, 19.7%, and 32.7% by history, by card and by history or card respectively.
Penta 3 coverage was 33.8%, 27.3% and 39% by history, by card and by history or card respectively.
Eighty three (28.8%) mothers answered that their children did not have polio birth dose vaccination
because they did not know the importance of vaccination while 57 (19.8%) mothers said services were not
available and 43 (14.9%) mothers responded that they did not know vaccination starts at birth. Other
responses included health facility was far (13.2%), service time was not convenient (12.8%), and did not
know place of vaccination (9.3%).
A total of 70 HEWs were interviewed in the nine study woredas . Forty one (58.6%) were female and 29
(41.4%) were male. Thirteen (18.6%), 4(5.7%), 5(7.1%) and 22 (31.4%) reported to have been trained in
Immunization in Practice (IIP), Integrated Refresher Training, Social Mobilization , and Newborn
Tracking respectively. Sixty (85.7%) HEWs reported to have received supervision of whom 39(63.9%) got
feedback.
Forty four (62.9%) HEWs reported that they conduct ANC while 27(38.6%) provide delivery services.
Thirty eight (54.3%) claimed to be registering births. A total of 71 CVSFPs were interviewed from the nine
study districts (woredas). Thirty five (49.3%) were female while 36 (50.7%) were male. Thirty nine (54.9%)
respondents reported that they could easily read and write, others 11(15.5%) could read and write with
difficulty and 21(29.6%) could not read or write at all. Thirty eight (53.5%), 21 (29.6%) and 16(22.5%) were
trained in community based surveillance, newborn tracking and social mobilization respectively.
Seventeen (23.9%) volunteers did not have any of the above training. Thirty forty seven (66.2%)
respondents reported that they know the number of pregnant women in the catchment areas and 36
(50.7%) registered births.
Forty seven TBAs were interviewed. The number of reported deliveries conducted by a TBA ranged from
2 to 40 during the previous 3-6 months. About one-fourth the TBAs said that they participate in
mobilization of the community during polio campaign whereas the rest did not participate in vaccination
vii
activities. None of the health center EPI coordinators knew Polio 0 coverage of their catchment area and
only 2 woreda EPI coordinators could provide figures on polio 0 coverage. Several participants in all
groups of FGDs mentioned that polio causes paralysis of legs and inability to walk. On the other hand,
they also mentioned symptoms that may not be indicative of polio. It was mentioned that there are
rumors that may discourage women and families from having their children vaccinated, although the
influence of these rumors was said to have decreased much nowadays.
Considering the findings from the different components of the study the following
recommendations were made: improving awareness of women, families and communities through
targeted IEC interventions; training and strengthening of supportive supervision; developing and
strengthening mechanisms for identification and follow up of pregnant women, use of ANC and
institutional delivery, birth registration and subsequent polio birth dose vaccination; designing and
strengthening strategies for improvement of accessibility and quality of maternal and newborn health
services.
1
INTRODUCTION
Health status of Ethiopian under five children
The most recent demographic and health survey (1) reported that the infant mortality rate is about 59
deaths per 1,000 live births. The estimate of child mortality is 31 deaths per 1,000 children surviving to 12
months of age, while the overall under-5 mortality rate for the same period is 88 deaths per 1,000 live
births. Although these figures show a major decline compared to the results of EDHS 2005 (2), the
country’s infant and under-five mortality rates remain very high. Moreover, there was no visible change
in neonatal mortality rates between EDHS 2005 (39/1000 live births) (2) and EDHS 2011 (37/1000 live
births) (1) reports.
Childhood immunization in Ethiopia
The Expanded Program on Immunization (EPI) was launched in Ethiopia in 1980 with the goal to
increase immunization coverage by 10% annually and achieve 100% DPT3 coverage by 1990 using three
service delivery strategies: fixed health units, outreach services and mobile team. The goal was not
achieved due to a combination of factors including; inadequate technical and managerial capacity, lack of
regular supervision and high staff turnover. During the last ten years a lot of effort has gone into building
the operational capacity of the immunization program in Ethiopia and according to the 2010/11 annual
report of the Federal Ministry of Health (3), Penta3 coverage reached 85%. However, in Afar and Somali
regions the reported coverage for the year was much lower (about 35%). A study conducted in
September,2009 by CORE Group Ethiopia (4) in its operational areas found Penta 3 coverage was about
68.4% with wide variation between 32% in Afder -Somali, 45% in Gambella, and 53% in Afar by history
or card. Similarly, Polio3 coverage was estimated at 68.6% coverage. Polio 0 coverage was lower (about
45%) than the other antigens. Of note is the sharp contrast with the EDHS 2011 report of 37% national
DPT3 coverage. Polio 0 coverage was reported at 20% in the EDHS 2011 (1). The reasons for the
discrepancies might need to be explored further.
Polio Eradication in Ethiopia
Polio eradication initiative in Ethiopia was started in 1996 based on the guideline provided by the World
Health Organization (WHO). Ethiopia has adapted the four strategies to eradicate polio. These are
achieving high routine immunization coverage, national supplemental immunization activities (SIAs),
acute flaccid paralysis surveillance and mop-up campaign. Since then the country has been conducting a
number of rounds of National Immunization Days (NIDs) and sub-National Immunization Days (NIDs)
to interrupt circulation of wild polio virus (WPV).
2
CORE Group Polio Project
The CORE Group is a voluntary network of 50 citizen supported private non-governmental organization
based in the USA. CORE was formed in 1997 and to date works in over 140 countries to promote and
improve primary health care. The main focus of the CORE Group is women and children in the context of
multi-sectoral development.
The CORE Group Polio Project (CGPP) was formed in 1999 and has been active participant in the global
Polio Eradication Initiative (PEI). It has been working in high risk areas of Angola, Bangladesh, Ethiopia,
India, Nepal and Uganda. Bangladesh, Uganda and Nepal have “graduated”. Currently the project is
functioning in Angola, Ethiopia and India with fund made available by USAID.
CORE Group Ethiopia started to function in November 2001 and has been supporting and coordinating
efforts of PVOs/NGOs involved in polio eradication activities. The CGPP National Secretariat staff
coordinates and ensures the quality of the social mobilization and community based surveillance
activities conducted by cadres of community-based volunteers. It closely collaborates with eight
international NGOs (CARE Ethiopia, Child Fund Ethiopia, Catholic Relief Service, Plan Ethiopia, Save the
Children USA, World Vision Ethiopia, International Rescue Committee and African Medical Research
Foundation) and four local NGOs (Pastoralist Concern, Harrerghe Catholic Secretariat, Alemtena
Catholic Church and Ethiopian Evangelical Church Mekane Yesus). In addition to these, CCRDA, WHO,
MOH and UNICEF are close allies of CORE Group Ethiopia.
Currently, CGPP Ethiopia works in 55 woreda1s in Somali (11 woredas), Amhara (4 woredas), Benshangul-
Gumuz (7 woredas), SNNP (8 woredas), Afar (6 woredas), Gambella (10 woredas) and Oromia (9 woredas)
regions of Ethiopia. In these regions, CGPP reaches a total of 4690972 of which 179795 are under one and
680042 are under five years old. In these woredas community Volunteers Surveillance Focal Persons
(CVSFPs) were trained and deployed at the village level to conduct house-house case detection and
reporting of AFP, Measles and NNT; mobilize community for polio SIAs and routine immunization
activities.
The current CORE Group Polio Project (CGPP) which extends from October, 2007 – September 2012 has a
goal of contributing to polio eradication by increasing population immunity and enhancing the
sensitivity of surveillance for AFP.
1 Woreda means district and is the most common term used in the Ethiopian literature and official documents
3
Figure 1: CORE Group Ethiopia’s project operations areas by woredas, 2012.
4
RATIONALE
CORE Group Ethiopia’s periodic reports, midterm assessment of CGPP and immunization surveys and
other studies have identified certain gaps in achieving CGPP objectives (4-6).
One such a gap is the very low coverage of OPV 0 (polio birth dose)2 in some CORE Group operational
areas. A major reason for the low OPV 0 coverage is the low coverage of institutional delivery in the
country in general, estimated at about 10% (1), and in pastoralist and semi-pastoralist areas in particular.
This low coverage deprives newborns the best opportunity of getting the newborn Polio 0 dose. Thus,
there is a need to devise mechanisms to identify newborns and be able to deliver Polio 0 (polio birth dose)
vaccination. A great potential lies in the combined activities of the CVSFPs and HEWs at the grass roots
level. CVSFPs, other community volunteers (CVs) and HEWs can create awareness among the
community about the importance of ANC, promoting facility delivery, newborn tracking and vaccination
of birth doses. This and other mechanisms have to be explored to identify newborns and vaccinate them
with Polio 0. This is also in line with recommendations of the Midterm CORE Group evaluation
conducted in Angola, Ethiopia and Utar Pradesh India to be able to play most effective role in polio
eradication (5)
This research is one of the three studies that have been identified as operations research areas for effective
implementation of CORE Group Ethiopia’s major activities. The other two are:-
AFP case detection and status of surveillance in pastoralist and semi-pastoralist communities of
CORE Group Polio Project implementation districts (woredas) in Ethiopia (7).
Cross Border Transmission of Wild Polio Virus (WPV) and Immunization Service Delivery in
CGPP Project Implementation International Border Areas in Ethiopia (8).
Figure 1 shows a conceptual frame work of the factors that affect Polio 0 (polio birth dose)
vaccination in pastoralist and semi-pastoralist areas of Ethiopia.
2 In this document Polio 0 and Polio birth dose are used interchangeably.
5
Figure 2: Conceptual Framework of Factors that Affect Polio 0 Vaccination in Pastoralist and Semi-pastoralist
Areas of Ethiopia
Individual level factors
Knowledge
Attitude
Socio-economic factors
Community level
factors
Informal/formal
influential groups
Values
Roads
Transport
Health system
factors
Types and level of
services
Access
Quality
Identification
of pregnant
women
Pregnancy
follow up
Child birth
Place
Attendant
Polio 0
vaccination
6
OBJECTIVES
General Objective
To examine pregnancy and child delivery practices and identify mechanisms for improving polio birth
dose coverage in CGPP implementation districts/ woredas.
Specific objectives
1. Assess identification, registration and follow up mechanisms of pregnant women in CGPP
implementation woredas
2. Identify places of child delivery and delivery attendant of women who gave birth in the previous
one year in CGPP implementation woredas
3. Assess the ways of OPV 0 vaccination delivery and their effectiveness in CGPP implementation
woredas
4. Suggest mechanisms for effective newborn tracking and OPV0 vaccination in the study areas
METHODS
Study design
A community based cross sectional study involving women who delivered in the previous one
year and facility based cross sectional study design involving key informant interviews of
community volunteers, Traditional Birth Attendants (TBAs), HEWs, program coordinators and
WHO surveillance officers were carried out. In addition, Focus Group Discussions (FGDs) were
conducted with community and religious leaders.
Study area
The study was conducted in CORE Group Ethiopia implementation pastoralist and semi pastoralist
project areas and included woredas (districts) in Afar, Benishangul, Oromia (Borena zone ), Gambella
and Somali regions.
The study areas have been identified through a consensus process of the CORE Group Ethiopia
secretariat using the following criteria:
1. Distance from center (Regional capital town)
2. Immunization performance (Coverage)
3. Cultural/ ethnic representation
4. Relevance to the study question
7
In using the above criteria, representation of worst and best scenarios was considered, while keeping in
mind feasibility, i.e. excluding extreme case of inaccessible and in secure areas.
Accordingly the following woredas (districts) were selected.
a. Gambella region : Larie, Gog3
b. Benishangul region : Kurmuk, Maokomo
c. Oromia region: Teltele
d. Afar region : Gewane
e. Somali region: Shinele, Filtu4, Dolobay5
Study populations
Women with recent deliveries
Women of reproductive age group (15-49 years) in the selected woredas of CGGP pastoralist and semi
pastoralist areas served as the source population. The study population was women who delivered in the
one year before data collection in the selected woredas.
Sample size determination
The sample size for the community based survey of women who delivered in the last one year was
calculated based on the single population proportion formula. The proportion used for the sample
size calculation was 45% which was the estimate for polio 0 coverage by CORE Group study in 2010
(6). The margin of error was put at 6% and confidence level at 95%. A design effect of 2 was
employed to account for variability due to cluster sampling. To account for non-response 10 % was
added.
n= 2 ( Zα/2)2 P (1-P) = 2 ( 1.96)2*(0.45)*(0.55) = 528
d2 (.06)2
Total sample size = 528 + 53 = 581
3 Replaced by Abol because of unforeseen security situations during the time of data collection
4 Replaced by Errer because of unforeseen security situations during the time of data collection
5 Replaced by Moyale because of unforeseen security situations during the time of data collection
8
Sampling Procedures
A multistage cluster sampling method with probability proportionate to the size (PPS) of the population
were employed to conduct the community based survey of women who delivered during the previous
one year.
Using the PPS technique 30 clusters were distributed among the nine selected woredas. The total number
of women who have delivered during the last one year per cluster (cluster size) is about 20 (581/30).
Studies have shown that a sample of 20 in a cluster of 30 clusters give a fairly adequate sample (9,10).
First the number of clusters to be included in a woreda were identified proportionate to the size of the
population and the corresponding number of women to be studied in a woreda were identified by
multiplying number of clusters per woreda by cluster size (twenty). The PPS technique is shown in Anex1.
The number of women who were interviewed by woreda is shown in the following table (Table 1). The
actual number of respondents was 600 due to rounding.
Table 1: Number of Clusters and Total Number of Women with Under one Children Required for the
Study in the Selected Woredas
Region/
Woreda
Population Eligibles Cumulative
Eligible
Sampling
Fraction
18282/30
=609
Random
number
49
Clusters
per
woreda
Samples
Per
Cluster
Total
samples
per
woreda
Gambella
Larie 35538 1174 1174 2 20 40
Gog (Abol) 18569 613 1787 1 20 20
Benshangul
Kurmuk 14989 555 2342 1 20 20
Maokomo 46415 1717 4059 3 20 60
Oromia
Teltele 76935 2924 6983 5 20 100
Afar
Gewane 34564 1071 8054 2 20 40
Somali
Shinele 113158 3847 11901 6 20 120
Filtu (Errer) 94847 3224 15125 5 20 100
Dolobay
(Moyale)
92860 3157 18282 5 20 100
Total 576736 18282 600
9
Data Collection
Questionnaire on attendance of antenatal care, place of delivery, delivery attendant, vaccination status of
the index child and other relevant variables were prepared in English. It was translated to Amharic and
Somali and back translated to ensure consistency. The questionnaire was pretested and administered by
trained interviewers.
Women who gave birth during the previous one year in the selected woredas were also a study population
for one of the other research topics mentioned above: AFP case detection and status of surveillance in
pastoralist and semi-pastoralist communities. In order to efficiently and effectively use resources, data
collection for the two studies was planned together, while separate proposals were developed for each.
Based on the sample size of women to be interviewed, 2- 10 interviewers each were selected to collect
data in each woreda for both studies (Table 2).
Partners at field level and health offices were contacted beforehand to make the necessary preparation for
data collection like selecting interviewers and providing technical, logistics and transportation support.
The interviewers had a minimum of diploma education, (experience in data collection preferable), spoke
fluently the local language, and were residents in the local area or vicinity. Data collection was
supervised by 2 supervisors in each study woreda. The supervisors had a minimum of a diploma
education and a previous experience in supervising community based data collection. They responded to
questions and queries of interviewers and corresponded with a coordinator and researchers whenever
necessary. The supervisors checked all filled questionnaires for completeness and consistency each day
before turning them to the coordinator. Job descriptions for the interviewers, supervisors and
coordinators were clearly spelt out and given to them in writing (Annex 2). A field guide manual was
developed for use by the interviewers and supervisors (Annex 3). Each interviewer was accompanied by
a community guide to help identify households and eligible respondents and facilitate communication
with the study population.
10
Table 2: Number of Interviewers, Supervisors, Field Guides by Woreda; and Coordinators by Region
Region/
Woreda
Clusters per
woreda
Total number of
women to be
interviewed
Interviewers Field
Guides
Supervisors* Coordinators
Gambella 1
Larie 2 40 4 4 2
Gog(Abol) 1 20 2 2 2
Benshangul 1
Kurmuk 1 20 2 2 2
Maokomo 3 60 6 6 2
Oromia 1
Teltele 5 100 9 9 2
Afar 1
Gewane 2 40 4 4 2
Somali 2
Shinele 6 120 10 10 2
Errer 5 100 9 9 2
Moyale 5 100 9 9 2
Total 30 600 55 55 18 6
The interviewers, supervisors and coordinators were trained for four days on general techniques of
interviewing and supervision and administration of each item in the questionnaire. Moreover, a pretest
was conducted in a selected pastoralist woreda before the final study began to assess the performance of
the study tools. Some revisions were made on the study instruments based on the feedback obtained
from the pretest.
In the selected woredas, a kebele6 was selected by simple random sampling among those that fulfilled the
inclusion criteria mentioned above. Some kebeles are divided into “gots” or villages. In such cases, one of
the villages was selected by simple random sampling procedure. Then, in the selected kebele/village a
central place was identified and a direction randomly identified (eg by spinning a bottle) to locate the
first household to start data collection. Data were collected in subsequent households until the end of the
selected direction is reached. If the selected household didn’t have eligible member then the nearest
household was included. If the allocated sample were not achieved, another direction was randomly
6 Kebele is the smallest administrative unit in Ethiopia
11
selected and data collection continued in a similar fashion until the required number of respondents was
obtained. In case eligible respondents were not available at the time of the survey a revisit (of no more
than 2 times) was arranged. If the required number of respondents were not obtained in one
kebele/village, another kebele/village was selected by using the simple random sampling method and
the procedure continued until the required sample size for the woreda was achieved.
Health Extension Workers (HEWs)
Key Informant Interviews were conducted with all HEWs in the selected for women’s interviews kebeles.
Interview guide questions including activities and services provided by the HEWs with special emphasis
to identifications of pregnant women, follow up, delivery attendance, vaccination status of children
including OPV0 and other relevant variables were prepared. The interviews were moderated by the
study supervisors, coordinates or research team members.
Community Volunteer Surveillance Focal Persons (CVSFPs)
All CVSFPS in the selected kebeles were included in the study. Interview guide questions including
activities undertaken by the CVSFPs, respondents’ knowledge and practice with regards to
immunization, OPV0 vaccination and surveillance were prepared. The interviews were moderated by
the study supervisors, coordinators or research team members.
Woreda and Health Center EPI Coordinators
A health center and a woreda health office that serves the catchment population of the selected woreda
were identified. In the selected health center and woreda office a staff member (usually known as EPI
coordinator) who is responsible for the immunization services was identified. Key informant interviews
were conducted according to a field guide which included birth identifications, antenatal care attendance,
delivery attendance, vaccination status of the index child and other relevant variables, possible
suggestions on how to identify pregnant women who had just given birth and mechanisms to deliver
OPV0.
Traditional Birth Attendants (TBAs)
TBAs who rendered pregnancy and delivery services were interviewed in the selected kebeles. They were
interviewed on pregnancy, child birth, postpartum care of women, and immunization status of children
and the mechanisms to reach children with OPV0 vaccination in their catchment area
Community and Religious Leaders
Focus Group Discussions (FGD) were conducted among community and religious leaders consisting of
6-8 men in each study kebele/woreda.
12
Guide questions were prepared to explore in-depth the knowledge, attitude, believes of group members
and the people they represent on newborn health and vaccination (OPV0), and maternal health service
utilization. FGD participants were people who were knowledgeable and able to express the opinions of
the community on the topic of discussion and were selected with the help of kebele, health staff and
partner organizations. The discussion took place in a “neutral” setting. The FGDs were conducted by
skilled/experienced moderators who had good knowledge of the subject of the study. This included
research team members and the study coordinators. The discussions were tape-recorded with the consent
of the participants and notes were taken by an assistant to the moderator.
Data Entry and Analysis
Quantitative data were entered and analyzed using SPSS version 17. Descriptive analysis included data
presentation using tables, graphs and appropriate summary figures.
Appropriate statistical tests (Chi squared test) and measures (OR, 95%CI) was used to asses significance
and strength of associations, respectively. Multiple regression analysis was used to measure the effect
different factors adjusted for possible confounders.
The records from these FGDs were transcribed in the language of the interview and then translated into
English for analysis. Data analysis was done using thematic approach.
The translated transcripts text files were copied into the “Open Code” computer program (ICT Services,
Umea University, 2006) for the study site under the same Project Title. After reading the transcripts
statement by statement and paragraph by paragraph, open coding of the texts was performed producing
substantive codes. As a number of substantive codes repeatedly came out across and between sites,
selective coding was performed where relevant codes were summarized to answer the thematic questions
Ethical Considerations
This is a cross sectional study mainly done to inform a program planning process and as such did not
need to go through a national IRB process. However, it was important to consult with the RHB and get
permission to undertake the survey from regional, woreda and kebele administrative authorities. Official
letters from the Regional Health Bureaus were written to the study sites as needed. Informed consent was
obtained from the study participants after explaining the purpose of the study. Participation of all
respondents in the study was strictly voluntary. During the training of interviewers, supervisors and site
coordinators emphasis was placed on the importance of obtaining informed consent. The interviewer
was made to sign on the consent form thereby verifying and taking responsibility of getting informed
consent.
13
RESULTS
Socio-demographic characteristics women who delivered in the previous one year
A total of 600 of women who delivered in the previous one year were included in the study. The socio-
demographic characteristics of the study population are shown in Table 3.
The mean age of the respondents was 26.3+ SD5.7, median 25 and range 15-49 years. Four hundred ninety
(81.7%) women could not read or write and 56 (9.3%) responded that they can read and write with
difficulty. The great majority of the women (98.2%) were currently married and about 79% had
monogamous marriage. Four hundred seven (78%) were Muslims. This was followed by different sects
of the Christian religion (16.8%) and Wakefeta (8.8%). Waketa is a religion observed in Oromia
Administrative Region. The majority ( 45.5%) of the respondent belonged to the Somali ethnic group
followed by Oromos. Four hundred ninety (81.7%) had different types of live stocks including camels,
cows, oxen, goat and sheep and 343(57.2%) owned some farm land. About 30% of the women responded
that they carry out income generating activities other than their main occupation, which is mainly cattle
rearing.
14
Table 3: Socio-demographic Characteristics of Women who Delivered a Baby in the Previous One Year
in Pastoralist and Semi-pastoralist Areas of CORE Group Polio Project Implementation Districts,
Ethiopia. 2012
Characteristics Number Percent
Region
Somali
Oromia
Benishagul
Gambella
Afar
Woreda
Shinele
Moyale
Errer
Teltele
Maokomo
Kurmuk
Lare
Abol
Gewani
320
120
100
100
100
80
60
20
60
40
20
40
40
53.3
20.0
16.7
16.7
16.7
13.3
10.0
3.3
10
6.7
3.3
6.7
6.7
Age
15-19
20-24
25-29
30-34
35-39
40-49
Don’t know
44
164
198
121
57
15
1
7.3
27.4
33.1
20.2
9.5
2.5
0.2
Literacy status
Can read and write easily
Can read and write with difficulty
Cannot read and write
54
56
490
9.0
9.3
81.7
15
Characteristics Number Percent
Grade Completed
None
1-6
7-13*
493
71
36
82.2
11.8
6.0
Marital status
Currently married
Divorced
Widowed
589
8
3
98.2
1.3
0.5
Type of marriage
Monogamous
Polygamous
I don’t know
464
123
2
78.8
20.9
0.3
Religion
Muslim
Protestant
Orthodox Christian
Catholic
Wakefeta
Others
407
78
15
5
53
9
72.8
13.0
3.0
0.8
8.8
1.5
Availability of livestock
Yes
No
490
110
81.7
18.3
Own farm land
Yes
No
343
257
57.2
42.8
Other income generating
Yes
No
I don’t know/missing
181
412
7
30.2
68.8
1.2
* Grade 13 means studied for one year after completing senior high school (ie Grade 12)
16
Pregnancy and child birth
Table 4 depicts reported number of pregnancies, deliveries and registered births.
The number of reported pregnancies ranged from 1-11. The mean number of pregnancies was 3.9+2.5 and
median 3.0. Ninety six women (16%) were pregnant for the first time while the majority (60.5%) were
pregnant 2-5 times. One hundred seven (17.8%) had delivered one child and 305(50.8%) delivered 2-4
times. The mean number of deliveries was 3.7 +SD2.2 and median 3.0. Of the last births 183( 30.5%) were
said to have been registered. One hundred (55%) were reported to be registered by HEWs. However,
only 85 (14.2%) were said to have been registered between birth and 14 days; 44 (52%) by HEWs, 11(13%)
by other CHWs and the rest by other health workers. Twenty two births (25.9%)who were registered
between birth and 14 days were registered at home, while 25 (29.4%) births registered at the health post
and 35 (41.2%) at health centers and hospitals.
Table 4: Pregnancy, Delivery and Birth Registration among Women who Delivered in the Previous
One Year in Pastoralist and Semi-pastoralist Areas of CORE Group Polio Project Implementation
Districts, Ethiopia. 2012
Pregnancy and child birth Frequency Percent
Number of Pregnancies 1 2-5 6- 11
96 363 141
16.0 60.5 23.5
Number of Deliveries 1 2-4 5-10
107 305 183
17.8 50.8 31.5
Last birth registered Yes No Don’t know
183 383 34
30.5 63.8 5.7
Last birth registered within 14 days Yes No/don’t know
85 515
14.2 85.8
Place last birth was registered within 14 days Home Health post Health center Hospital Other/unspecified
22 25 20 15 3
25.9 29.4 23.5 17.6 3.5
17
Maternal Health Service Utilization
Tables 5-7 show maternal health service utilization by the study population during the last pregnancy
and delivery. Three hundred twelve women (52.0%) had attended antenatal care at least once during the
last pregnancy. The mean number of months at start of ANC was 4.5+ SD1.5 and median was 5.0 months.
The majority (51% of those who attended ANC) started attending during the second trimester (4-6
months). The mean number of ANC visits was 3.1+ SD 1.5 and median 3.0. One hundred fifteen women
(19.2%) had four or more ANC visits. Of the 312 women who had ANC 180 (57.7%) attended ANC for
check up while 121 (38.8%) attended ANC because of a health problem. About 48% of the women who
had attended ANC attended in health centers and 34% in health posts. It was reported that the decision
to attend ANC was mainly made by the respondent (67.2%) or both ( 15.3%) the respondent and the
husband.
Table 5: Antenatal Care Attendance during the pregnancy of the index child in pastoralist and semi-
pastoralist areas of CORE Group Polio Project Implementation Districts, Ethiopia. 2012
Antenatal Care Number Percent
Had Antenatal care Yes No
312 288
52.0 48.0
Time when antenatal care started First trimester (1-3 months) Second trimester (4-6 months) Third trimester (7-9 months) Does not remember time Did not attend ANC
94 158 47 13 288
15.7(30.1)* 26.3(50.6) 7.8 (15.1) 2.2(4.2) 48.0
Number of ANC visits None 1-3 visits 4 and above
288 197 115
48.0 32.8 19.2
Reason for ANC (n=312) ANC check up Health problems Other/unspecified
180 121 11
57.7 38.8 3.5
Place of ANC attendance (n=312) Health post Health center Government hospital Private clinic Private hospital
106 149 24 21 5
34.0 47.8 7.7 6.7 1.6
Decision maker on ANC (n=312) Women Husband Both Others/unspecified
209 39 48 16
67.2 12.4 15.3 5.1
* Percent in bracket indicates proportion of those who attended ANC
18
Five hundred forty nine women (91.5%) delivered their last baby at home, while 45(5.5%) delivered in
health centers and hospitals and the majority (80.0%) of the attendants at home were untrained or trained
traditional birth attendants. The HEWs attended only 9(1.6%) births and five (0.8%) women delivered in
health posts (Table 6).
In the majority of the cases (79.5%) the respondents decided on the place of delivery, where as husbands
alone were reported to have decided in only 7% of the cases.
Table 6: Delivery Care During the Last Birth in Pastoralist and Semi-pastoralist Areas of CORE
Group Polio Project Implementation Districts, Ethiopia. 2012
Place of delivery Home Health post Health center Hospital Other
549 5 18 27 4
91.5 0.8 3.0 4.5 0.7
Decision on place of delivery Women Husband Both Mother-in-law TBA Others
477 42 39 11 12 16
79.5 7.0 6.5 1.8 2.0 2.7
Delivery attendant (n=549) TBA TTBA HEW CVSFP Neighbor Others
337 102 9 5 93 3
61.4 18.6 1.6 0.9 16.9 0.5
Figure 3 shows the reasons for home delivery. These reasons include familiar birth attendants (46.4%),
distance to health facility (24.4%), unavailability of transport (16.0%), delivery in health facility not
culturally encouraged (10.8%) and non-friendly health services (8.9%). Twenty four women (4.4%)
answered that the health facility was not functioning at the time they had labor.
19
Figure 3: Reasons for home delivery of the last child in pastoralist and semi-pastoralist areas of CORE Group Polio
Project Implementation Districts, Ethiopia. 2012
Two hundred seventy nine (46.5%) women reported to have attended postnatal care (PNC). Seventeen
(2.8%) attended PNC within the first 24 hours, while 38(6.3%) visited health facilities between 24 hours
and seven days. A relatively high number of the respondents, 198(33%) visited health facilities at the 45th
day after delivery. Twenty one women (3.5%) reported to have postnatal care after 45 days of delivery.
The main reason of post natal care visit was immunization of children ( 84.6% of those who had PNC)
followed by maternal sickness (13.9 %). Maternal health check up was aimed at only 5.4% of the cases
(Table 7)
46.4
24.4
16
10.8
8.9
4.4
4
3.1
2.7
2.2
1.1
0.7
0 10 20 30 40 50 60 70 80 90 100
0 50 100 150 200 250
Familiar attendants
Healt faility far
No transport to health facility
Delivery in HF not culturally encouraged
Non-friendly services
Health facility not functioning
Labor was light
Husband's decision
Labor was too fast
Health facility expensive
Labor took place at night
Being not sick
Percentage
Frequency
20
Table 7: Postnatal Care after Last Delivery in Pastoralist and Semi-pastoralist Areas of CORE Group
Polio Project Implementation Districts, Ethiopia. 2012
Attended post natal care Yes No
279 321
46.5 53.5
Time of first PNC attendance (n=279) Within 24 hours Between 24 hours and 7 days PNC at 45 days PNC after 45 days Unspecified
17 38 198 21 5
6.1 13.6 71.0 7.5 1.8
Place of first PNC attendance (n=279) Health post Health center Government hospital Private clinic Private hospital
141 99 19 17 3
50.5 35.5 6.8 6.1 1.1
Purpose of PNC attendance*(n=279) Maternal check up Child immunization Mother’s sickness Child check up Child sickness
15 236 39 12 20
5.4 84.6 14.0 4.3 7.2
* Multiple answers may be given
Association of different factors with Antenatal Care Utilization
In bivariate analysis, literacy status, religion, availability of live stock, carrying out other income
generating activities were significantly associated with ANC attendance, whereas age group, owning
farm land and number of pregnancies were not. When adjusted for other factors, religion having live
stock, and income generating activities were significantly associated with ANC attendance at P<0.05.
Those who had live stocks had a 60% as much less chance to attend ANC, compared to those who do
not have, while those who had other income generating activities had about twice the chance of attending
ANC compared to those who had not . Regarding religion, those who follow the Wakefeta religion had a
very low ANC attendance rate (7.5%) and the odds of attending ANC was 6% that of the Muslims. Those
who could read and write easily had about twice the chance of attending ANC in bivariate analysis, but
the association turned to be non-significant when adjusting for other factors (Table 8).
21
Table 8: Association of Socio-demographic Characteristics with Antenatal Care Attendance in
Pastoralist and Semi pastoralist Areas of CORE Group Polio Project Implementation Districts,
Ethiopia. 2012
Characteristics Had ANC Yes No
OR 95%CI
Adjusted OR 95%CI
Age group 15-19 20-34 35-49
29(65.9) 246(50.9) 37(51.4)
15(34.1) 237(49.1) 35(48.6)
1.00 0.54(0.27,1.07) 0.55(0.23,1.27)
Literacy status
Can read and write easily Reads and writes with difficulty or unable to read and write
37(68.5%) 275(50.4%)
17(31.5%) 271(46.9%)
2.1(1.2, 4.1)* 1.00
1.9 (0.97, 3.63)
Religion Muslim Christian Wakefeta
255(58.4) 49(47.5) 4(7.5)
182(41.6) 52(51.5) 49(92.5)
1.00 0.67(0.43,1.06) 0.06(0.02,0.17)
0.65(0.41,1.02) 0.06(0.02,0.17)
Any livestock Yes No
239(48.8) 73(66.4)
251(51.2) 37(33.6)
0.48(0.3, 0.7)* 1.00
0.63(0.40,1.00)*
Own farm land Yes No
175(51.0) 137(53.3)
168(49.0) 120(46.7)
0.92(0.66,1.26)
1.00
Other income generating activities Yes No
113(62.4) 195(47.3)
68(37.6) 217(52.7)
1.8(1.3, 2.7)* 1.00
1.89(1.28,2.79)*
Number of pregnancies 1 2-5 6 and above
53(55.2) 185(51.0) 74(52.5)
43(44.8) 178(49.0) 67(47.5)
1.16(0.67,2.01) 0.94(0.63,1.42) 1.00
* P<0.05
22
Factors Associated with Home Delivery
As mentioned earlier, a large majority of the deliveries took place at home. Table 9 shows the
relationship of socio demographic factors with home delivery. Inability to read and write, , religion,
availability of livestock, unavailability of other income generating activities, higher number of
pregnancies, and lack of ANC attendance were statistically significantly associated with home delivery.
The binary logistic regression analysis results presented in Table 5 show adjusted values without (normal
Font) and with inclusion of ANC (Italic Font). After adjusting for other factors, Wakefeta Religion and no
other income generating activities were statistically significantly associated with home delivery when
ANC was not included in the model. When ANC is included, none of the variables retained statistical
significant. The odds of attending ANC by those who delivered at home was 19% of those who did not
attend ANC.
Table 9: Association of Different Factors with Home Delivery of the Last Child in Pastoralist and Semi
pastoralist Areas of CORE Group Polio Project Implementation Districts, Ethiopia. 2012
Characteristics Delivery home Yes
No
OR 95%CI
Adjusted OR 95%CI
Age group 15-19 20-34 35-49
38(86.4) 442(91.5) 68(94.4)
6(13.6) 41(8.5) 4(5.6
1.0 1.70(0.61,4.53) 2.68(0.62,12.23)
Literacy status Can read and write easily Reads and writes with difficulty or unable to read and write
45(83.3) 504 (92.3)
9(16.7) 42(7.7)
0.42(0.19,0.91)* 1.00
0.71(0.29, 1.73) 0.86(0.35,2.14)7
Religion Muslim Christian Wakefeta
408(93.4) 81(80.2) 52(98.1)
29(6.6) 20(19.8) 1(1.9)
1.00 0.29(0.15,0.56)* 3.70(0.59,74.4)
3.64(0.48,27.74) 11.74(1.51,91.61) 1.43(0.18,11.66) 5.71(0.70, 46.83)
Any livestock Yes No
454(92.7) 95(86.4)
36(7.3) 15(13.6)
2.0(1.00, 3.90)* 1.00
1.7(0.88,3.56) 1.70(0.84,3.46)
Other income generating activities Yes No
156(86.2) 386(93.7)
25(13.8) 26(6.3)
0.42(0.24,0.75)* 1.00
0.53 (0.29,0.98)* 0.62(0.33,1.17)
7 Italic font ANC included in the model
23
Characteristics Delivery home Yes
No
OR 95%CI
Adjusted OR 95%CI
Number of pregnancies 1 2-5 6 and above
82(85.4) 332(91.5) 135(95.7)
14(14.6) 31(8.5) 6(4.3)
0.26(0.09,0.76)* 0.48(0.17,1.19) 1.00
0.77(0.37, 1.64) 0.41(0.14,1.17) 0.77(0.36,1.66) 0.40(0.13, 1.18)
Had ANC Yes No
270(86.5) 279(96.9)
42(13.5) 9(3.1)
0.21(0.10,0.43) 1.00
0.19(0.09,0.44)*
*P<0.05
Factors Associated with Registration of the Last Birth within 14 days after delivery
Table 10 presents the association of socio-demographic factors with registration of last birth. In bivariate
analysis, followers of Christian Religion and Wakefeta Religion were highly likely and less likely
respectively, to be associated with birth registration compared to Muslims (P<0.05). Families with
additional income generating activities had a higher chance of last birth registered (OR 1.8(1.1, 2.9) in
bivariate analysis and adjusted for other factors, when ANC is not included in the model (OR
1.75(1.07,2.86). Those who had 2-5 number of pregnancies had statistically significant association with
registration of births within the first 14 days, but number of pregnancies did not have significant
association when adjusted for other factors. As in the case of deliveries none of the other factors had
significant association with birth registration when ANC is included in the model (Italic Font).
As expected, home delivery was significantly and inversely associated with lower chance of association
with birth registration (Adj OR 0.76 (04, .14), not shown in the table)
24
Table 10: Relationship Between Socio-demographic Factors and Registration of Births within 14 Days
After Delivery in Pastoralist and Semi pastoralist Areas of CORE Group Polio Project Implementation
Districts, Ethiopia. 2012
Characteristics Last birth registered within 14 days of delivery Yes No
OR 95%CI
Adjusted OR 95%CI
Age group
15-19 20-34 35-49
5(11.4) 67(13.9) 12(16.7)
39(88.6) 416(86.1) 60(83.3)
1.0 1.26(0.47,4.23) 1.56(0.46,5.56)
Literacy status
Can read and write easily Reads and writes with difficulty or unable to read and write
12(22.2) 72(13.2)
42(77.8) 474(86.8)
1.8(0.95,3.7) 1.00
1.54(0.73,3.25)8 0.59(0.28,1.24)
Religion Muslim Christian Wakefeta
59(13.5) 22(21.8) 2(3.8)
378(86.5) 79(78.2) 51(96.2)
1.00 1.78(1.00,3.18*) 0.25(0.03, 1.00)*
0.47(0.11, 2.06) 0,23(0.50,1.07) 1.79(1.02,3.16)* 0.28(0.07,1.17)*
Any livestock Yes No
69(14.1) 15(13.6)
421(85.9) 95(86.4)
1.04(0.55,1.98) 1.00
Own farm land Yes No
49(14.3) 35 (13.6)
294 (85.7) 222(86.4)
1.06(0.65,1.73) 1.00
Other income generating activities Yes No
35(19.3) 48(11.7)
146(80.7) 364(88.3)
1.8(1.1, 2.9)* 1.00
1.54(0.93,2.54) 1.75(1.07,2.86)
Number of pregnancies 1 2-5 6 and above
23(8.0) 34(17.3) 27(23.5)
265(92.0) 163(82.7) 88(76.5)
0.28(0.15,0.54)* 0.68 (0.37,1.25) 1.00
0.98(0.49,1.96) 0.62(0.28,1.34) 0.64(0.30,1.38) 0.65(0.37,1.34)
8 Blue font and italics ANC included in the model
25
Characteristics Last birth registered within 14 days of delivery Yes No
OR 95%CI
Adjusted OR 95%CI
Had ANC
Yes
No
61(19.6)
23(8.0)
251(80.4)
265(92.0)
2.54(1.47.4.38)*
1.00
2.59(1.49, 4.47)*
*P<0.05
Knowledge of vaccine preventable diseases and vaccination status of children
As shown in figure 4, three hundred five (50.8%) women said they know polio. This is followed by
measles, 268(44.7%) tuberculosis 147(37.0%) and Tetanus 119(24.8%).
Regarding the time when childhood vaccination starts 174 (29.0%) women answered that it starts at birth,
while 156 (26.1%) said it starts at six weeks after birth and 330(55%) said they don’t know when
vaccination starts.
Figure 4: Reported knowledge of vaccine preventable diseases by women who gave birth in the previous one year in
pastoralist and semi pastoralist areas of CORE Group Polio Project Implementation Districts, Ethiopia. 2012
50.8
44.7
24.5
24.8
19
21.6
8.5
0 10 20 30 40 50 60 70 80 90 100
0 50 100 150 200 250 300 350
Polio
Measels
Tuberulosis
Tetanus
Pertussis
Pneumonia
Diphteria
Percent
Frequency
26
Vaccination status of children born in the previous one year
Table 11 shows vaccination status of children born in the last one year by history, card and both history
and card. BCG coverage was 53.7%, 40.8%, 57.2% by history, by card and by history or card respectively.
Polio 0 coverage was 29.7%, 19.7%, and 32.7% by history, by card and by history or card respectively.
Penta 3 coverage was 33.8%, 27.3% and 39% by history, by card and by history or card respectively.
Table 11: Vaccination Status of Children Born in the Previous One Year in Pastoralist and Semi
pastoralist Areas of CORE Group Polio Project Implementation Districts, Ethiopia. 2012
Antigens/vaccines Vaccinated by history
No (%)
Vaccinated by card
No (%)
Vaccinated by history or card
No (%)
BCG 322(53.7) 240(40.8) 343(57.2)
Polio 0 178(29.7) 118(19.7) I96(32.7)
Polio 1 324 (54.0) 258(43.0) 357(59.5)
Polio 2 269(44.8) 212(35.3) 325(54.2)
Polio 3 196 (32.7) 152(25.3) 222(37.0)
Penta 1 349(58.2) 274 (45.7) 379(63.2)
Penta 2 273 (45.5) 224(37.3) 304 (50.7)
Penta3 203(33.8) 164(27.3) 234 (39.0)
Measles 102 (17.0) 86(14.3) 126(21.0)
PCV1 155 (25.8) 141(23.5) 178(29.7)
PCV 2 72(12.0) 65(10.8) 80(13.3)
PCV3 55(9.2) 51(8.5) 59(9.8)
As shown in Table 12, religion and attending ANC or birth registration within 14 days of delivery (Italic
Font) were statistically significantly associated with polio birth dose vaccination. ANC (Black Font) and
birth dose registration (Italic Font) were alternatively included in the model.
27
Table 12: Factors Associated with Polio Birth Dose Vaccination in Pastoralist and Semi pastoralist
Areas of CORE Group Polio Project Implementation Districts, Ethiopia. 2012
Characteristics Polio birth dose vaccinated Yes No
OR 95%CI
Adjusted OR 95%CI
Age group 15-19 20-34 35-49
11(25.0) 163(33.7) 22(30.6)
33(75.0) 320(66.3) 50(69.4)
1.53(0.72,3.30) 1.32(0.52,3.36)
Literacy status Can read and write easily Reads and writes with difficulty or unable to read and write
22(40.7) 174(31.9)
32(59.3) 373(68.1)
1.47(0.83,2.60)
Religion Muslim Christian Wakefeta
144(33.0) 46(45.5) 4(7.5)
293(67.0) 55(54.5) 49(92.5)
1.70(1.07,2.70)* 0.17(0.05,0.49*)
1.81(1.16,2.84)* 0.20(0.07,0.58)* 1.57(0.98,2.53) 0.19(0.65,0.54)*
Any livestock Yes No
157(32.0) 39(35.5)
333(68.0) 71(64.5)
0.85(0.56,1.33)
Own farm land Yes No
112(32.7) 84(32.7)
231(67.3) 173(67.3)
1.00(0.70,1.43)
Other income generating activities Yes No
68(37.6) 126(30.6)
113(62.4) 286(69.4)
1.37(0.95,1.97)
1.27(0.86,1.87) 1.21(0.81,1.81)
Number of pregnancies 1 2-5 6 and above
28(29.2) 123(34.6) 43(30.5)
68(70.8) 233(65.4) 98(69.5)
0.94(0.51,1.72) 1.07(0.69,1.65)
Attended ANC Yes No
120(38.5) 76(26.4)
192(61.5) 212(73.6)
1.74(1.23,2.47)*
1.48(1.02,2.14)*
Birth registered within 14 days Yes No
61(72.6) 135(26.2)
23(27.4) 381(73.8)
7.5(4.5, 12.6)*
6.35(3.74,10.77)*
28
Reasons for not having polio birth dose vaccination
As shown in figure 5, 83 (28.8%) mothers answered that their children did not have polio birth dose
vaccination because they did not know the importance of vaccination. Fifty seven (19.8%) mothers said
services were not available and 43 (14.9%) mothers responded that they did not know vaccination starts
at birth. Other responses included health facility was far (13.2%), service time was not convenient
(12.8%), did not know place of vaccination (9.3%), mother was too busy, delivery took place not in the
normal (current residence area), fear of side effects, bad health workers approach, and mother can’t go
out of home soon after delivery.
Figure 5: Reasons for not vaccinating children born in the previous one year is pastoralist and semi-pastoralist areas
Doesn't know importance of
vaccination
Service not avaialble
Does 't know vaccination
starts at borth
Health facility far
Service time not conveniet
Doesn't know place of
vaccination
Mother too busy
Delivery occurred elsewhere
Fear of side effects
Bad health workers
approach
Mother can't go out
Long waiting time
Religious reasons
28.8
19.8
14.9
13.2
12.8
9.4
5.9
5.6
3.5
3.1
2.1
1.7
0.1
0 10 20 30 40 50 60 70 80 90 100
0 10 20 30 40 50 60 70 80 90
Percent
Frequency
29
Knowledge on prevention, transmission and vaccination against polio
Three hundred seventeen(52.8%) respondents answered that polio is prevented by vaccination and
118(19.7%) said it is prevented by personal and environmental hygiene, while 245(40.8%) said they don’t
know. Polio was said to be transmitted by sneezing/cough by 68 (13.1%), open defecation 58(9.7%),
contaminated water 51(8.5%) respondents while 445(74.2%) said they don’t know how it is transmitted.
The time when polio vaccination starts was indicated to be birth or within 2 weeks by 220 (37.4%) women
and 281(46.7%) answered that they don’t know or did not tell the time when polio vaccination starts
(Table 13).
Table 13 : Knowledge on Prevention, Transmission and Polio Vaccination in Pastoralist and Semi
pastoralist Areas of CORE Group Polio Project Implementation Districts, Ethiopia. 2012
Knowledge on polio prevention and transmission and time vaccination starts
Frequency Percent
Polio prevention * Vaccination Personal environmental Hygiene Others** Don’t know
317 118 16 245
52.8 19.7 2.7 40.8
Polio Transmission Sneezing/cough Open defecation Contaminated water Body fluids Others Don’t know
68 58 51 30 16 445
13.1 9.7 8.5 5.0 2.7 74.2
Time to start polio vaccination At birth At two weeks At one month Don’t know/not specified
163 61 95 281
27.2 10.2 15.8 46.7
* - Multiple answers possible
** - Others include use of bed nets, not using the same eating utensils, clothes, not playing with patient,
Findings of interviews with HEWs, CSFP, TBAs and EPI coordinators
A total of 70 HEWs were interviewed in the nine study woredas. The median age of the respondents was
24.0 years. It ranged from 18-45 years. Forty one (58.6%) were female and 29 (41.4%) were male. The
mean (and median) duration of work of the respondents was 3.0 years and ranged from 3 months to 6
years. In the majority of the cases 59(84.3%) the health posts were located in rural areas. Thirteen (18.6%),
30
4(5.7%), 5(7.1%) and 22 (31.4%) reported to have been trained in IIP, Integrated Refresher Training, Social
Mobilization, and Newborn Tracking respectively. Sixty (85.7%) HEWs reported to have received
supervision of whom 39(63.9%) got feedback.
Fifty one (72.9%) HEWs claimed to know the number of pregnant women in their kebeles, using several
means including house to house visit , estimating from the kebele population proportion (usually 3.4%)
and getting numbers from the woreda health office. Forty four (62.9%) HEWs reported that they conduct
ANC while 27(38.6%) provide delivery services. Thirty eight (54.3%) claimed to be registering births.
Regarding actions to improve Polio 0 coverage, the HEWs recommended to give health education so that
women bring their children to the health facility for vaccination 23(32.9%); delivering Polio 0 house to
house 20(28.6%); mother to deliver in health institutions ; and equip health posts with fridge, vaccines
and supplies 8(11.4%) each. Others suggested a combination of the measures mentioned above.
A total of 71 CVSFPs were interviewed from the nine study districts (woredas). The mean age of the
respondents was 33.9 + 9 and median 30 years. Thirty five (49.3%) were female and while 36 (50.7%) were
male. Thirty nine (54.9%) respondents reported that they can easily read and write, others 11(15.5%)
could read and write with difficulty and 21(29.6%) could not read or write at all. The mean duration of
work was 3.4 years + 2.4, and median was 3.0 years.
Thirty eight (53.5%), 21 (29.6%) and 16(22.5%) were trained in community based surveillance, newborn
tracking and social mobilization respectively. Seventeen (23.9%) did not have any of the above training.
Thirty nine (54.9%) of the CVSFPs reported to have received supervision during the last six months, of
whom 22(59.5%) got feedback. On the other hand, forty seven (66.2%) respondents reported that they
know the number of pregnant women in the catchment areas and 36 (50.7%) registered births.
Regarding actions to improve vaccination of polio birth dose, 32 (45.1%) CVSFPs suggested educating
women to give birth at health facilities and/or bring the newborn for Polio 0 vaccination. Fifteen (21.1%)
recommended house to house Polio 0 vaccination, and 8 (11.3%) to assign personnel and equip the health
posts with the necessary facilities and supplies.
Interview was conducted with 47 TBAs in the 9 districts. The number of reported deliveries conducted
by the TBAs ranged 2-40 during the previous 3-6 months. Concerning how they identify pregnant women
and conduct deliveries, most reported that it is easy because the abdomen shows up, the presence of
symptoms like loss of appetite and tiredness in the absence of menstruation indicates pregnancy and
many are contacted by women for advice and reassurance about the position of the fetus. A few were
only called when labor starts. Two TBAs mentioned that they were no more allowed to follow pregnant
women or conduct deliveries on their own. Only four TBAs knew that the first dose of polio is given at
31
birth or within 14 days after birth. About one-fourth of the TBAs said that they participated in
mobilization of the community during polio campaign whereas the rest did not participate in vaccination
activities. Concerning improvement of polio birth dose vaccination, their suggestions included house to
house vaccination, since women cannot go out of home during the first 15 days, particularly according to
Somali culture; ANC and advice on Polio birth dose vaccination; opening and staffing a health post in
the village; house to house visit to identify and follow pregnant women, educate women, families and
elders about the first (birth) polio dose; delivery in health facilities; closer ties between TBAs and health
workers; and informing HEWs/health workers about deliveries as soon as they occurred.
Eleven health center and 9 Woreda (district) EPI coordinators were also interviewed. None of the health
center EPI coordinators knew Polio 0 coverage of their catchment area. They also did not know ANC and
Institutional delivery coverage. Concerning improving polio birth dose coverage, their suggestions
included increase institutional delivery, educate the community on polio birth dose, house to house
vaccine delivery by HEWs, and provision of polio birth doses at outreach sites, in that order. Only 2
woreda EPI coordinators could provide figures on polio 0 coverage. Regarding suggestions to improve
Polio 0 coverage, they mentioned house to house search and vaccination by HEWs, institutional delivery,
HEWs identify pregnant women through house to house assessment and link to health facility, avail
refrigerators, supplies and vaccines at all health facilities, provide outreach services and networking with
TBAs and volunteers .
Findings from FGD with community leaders
Participants mentioned that polio is known by different names and indicated symptoms and signs of the
disease. One such sign related name can be translated as “death of legs”. Several participants in all
groups mentioned that polio causes paralysis of legs and inability to walk. On the other hand, they also
mentioned symptoms that may not be indicative of polio such as leg swelling and fainting. Some
participants attributed curse as a cause of polio while several answered that they don’t know the cause of
polio. Concerning transmission of polio, some FGD participants mentioned a number of ways including
airborne from infected person, close contact with a polio patient, body contact with a polio patient and
the like. A few mentioned that polio is transmitted by movement of people and migration from
neighboring countries. Several participants said that they do not know the ways of polio transmission.
Regarding prevention, many participants mentioned vaccination. However, in some cases correct specific
knowledge may be lacking on polio vaccination. An example of such a response is that polio is prevented
by vaccination of pregnant women and children. Other responses include not eating together, not sharing
clothes, exclusive breast feeding, feeding children with a lot of milk and protecting children from dirt and
keeping them clean.
32
Concerning reasons that encourage or discourage immunization, almost all participants agreed that there
is no reason related to religion that discourages immunization in the study areas. It was mentioned that
there are rumors that may discourage women and families from having their children vaccinated,
although the influence of these rumors has decreased much nowadays. Some participants said that they
may still exist to a greater extent in remote areas. The rumors include sterility of women following
vaccination, children may be disabled after vaccination and vaccination causes a lot of suffering and pain
in children. An instant was mentioned where a health care provider had to sign an agreement with
women declaring that vaccination would not cause sterility. Actual and perceived side effects were
another reason for avoiding vaccination. These include fever, excessive crying, swelling at the site of
vaccination, diarrhea and vomiting. Side effects are said to particularly influence subsequent rounds of
vaccinations if they occur during the first round.
The measures cited by the participants to improve polio birth dose were teaching about and promoting
institutional delivery, registering pregnant women and newborns by HEWs, house to house polio birth
dose vaccination by HEWs, and involving elders and religious leaders in promoting newborn
vaccination. The importance of creating mechanisms for kebele administration to take responsibility for
birth dose vaccination, using all available venues such as mosques, churches and market places to inform
about polio birth dose was emphasized by some participants.
DISCUSSION
Six hundred women were included in the study making the response rate 100%, although non-response
to few questions was noted and that the variables had to be discarded from analysis. A 100% overall
response was possible because in case of absence or non-response, data collection continued until the
allocated sample size was obtained. Absence or non -response that needed to be replaced was noted in
only 12 (2%) cases. Data not included in the results were related to economic status assessment such as
number and type of cattle and land size since they were found not to be meaningful and or the majority
did not respond to them. This could be because of lack of information by the women (only the husband
knows) or fear of implications of telling the truth, although the purpose of the study was well explained
at each study site.
The socio-demographic characteristics of the study population generally represented the distribution of
pastoralist and semi-pastoralist population in CORE Group Ethiopia Project Areas. The majority of the
respondents (91%) were either unable to read and write or did so with difficulty. While this is expected in
such relatively disadvantage places, it has important implications for planning, organizing and provision
of proper IEC (Information, Education & Communication). Health education messages and methods
should be geared accordingly. The great majority of the respondents (98.2%) were married reflecting the
33
expectation that child bearing generally takes place in the context of marriage in the study area. On the
other hand about one fifth of the current marriages were polygamous. No difference was however noted
in maternal health service utilization or polio birth vaccination in this study between polygamous and
monogamous marriages.
Only 85 (14.2%) of the births in the previous one year were said to have been registered between birth
and 14 days, 44 (52% by HEWs). Interview with HEWs also indicated that 38 (54.3%) claimed to be
registering births. Thus, a substantial number of births go unregistered between birth and 14 days
making the likelihood of receiving polio birth dose highly unlikely.
Three hundred twelve women (52.0%) had attended antenatal care at least once during the last
pregnancy. The most recent demographic and health survey (1) reported a relatively lower attendance
rate for some of the regions the study population is located in; Afar (35.3%), Somali (25.3%), Gambella
(57.7%) , Benishangul Gumuz (40.8%) and Oromia (39.5%). Differences in coverage were also noted from
the annual report of the Federal Ministry of Health (11) ; Afar (26.4%), Somali 30.9%, Gambella (39.1%),
Benishagul Gumuz (52.6%). The differences may be explained by the fact that the study woredas may not
be a typical representative of the administrative region in which it is located. It may also be the case that
the activities of CGPP and partners’ projects may have improved coverage in some instances.
Five hundred forty nine women (91.5%) delivered their last baby at home, while 45(5.5%) delivered in
health centers and hospitals. And the majority (80.0%) of the attendants at home were untrained or
trained traditional birth attendants. The HEWs attended only 9(1.6%) births. The very high proportion of
home delivery and unskilled delivery attendant even by Sub-Saharan African standard is similar to the
reports of the recent results of the Ethiopian Demographic Health Survey (1), although there may be
some variations among and between the administrative regions of the country. The report of the FMOH
gave a relatively higher result for the country and some regions (11). As indicated earlier, this
persistently low institutional delivery coverage becomes a major constraint to deliver the polio birth dose.
The reasons for delivering at home include familiar birth attendants (46.4%), distance to health facility
(24.4%), unavailability of transport (16.0%), delivery in health facility not culturally encouraged (10.7%)
and non-friendly health services (8.9%). Twenty four women (4.4%) answered that the health facility was
not functioning at the time they were in labor. Thus the decision to deliver at home and not health
facilities appear to mainly be related to health service delivery factors while cultural or community
related reasons seem to contribute to a lesser extent.
Decision making about place of delivery and ANC attendance was made by the women in more than 65%
of the cases in this study population. This is different from the general finding of the Ethiopian
34
Demographic and Health Survey which reported that health care of women is mainly decided jointly
61 % or by the husband 25% and to a lesser extent by the wife alone (13.4%) (1).
About 46% of the respondents reported to have attended postnatal care. However, a relatively high
number of the respondents, 198(33%) visited health facilities at the 45th day after delivery. Twenty one
women reported to have postnatal care after 45 days of delivery. Thus, for most of the women the chance
of their babies receiving polio birth dose during post natal care is rare. It is also worth noting that the
main reason of post natal care visit was immunization of children. While this shows the relative
importance given to childhood immunization and has the potential for improving coverage, it shows that
post natal maternal health is not getting due priority.
When adjusted for other factors, religion, having live stock, and income generating activities were
significantly associated with ANC attendance (P<0.05) in both bivariate and multiple logistic regression
analyses. Regarding religion, those who follow the Wakefeta religion had a very low ANC attendance
rate (7.5%) which entails that a lot has to be done concerning the followers of the religion to improve
ANC overage as ANC is found to be the most important determinant of institutional delivery, birth
registration and polio birth dose immunization (Please see below). Having live stock was inversely
associated with ANC care. It is possible that this is related to the pastoralist nature of the population
with livestock, where people travel with cattle in search of grazing land and water, and would not have
access to ANC services.
The odds of engaging in other income generating activities were 1.8 times higher among the ANC
attendees compared to non-attendees. A possible explanation is that other income generating activities
may enable the family to stay around towns and hence improves the chance of ANC attendance.
Being able to read and write easily did not maintain significant association with ANC attendance when
adjusted for other factors. It is possible that the level of literacy that was defined as being able to read and
write easily and its effects on other determinants of health care is not good enough to make the required
changes or differences in the study setting. The population appears fairly homogeneous in this respect.
After adjusting for other factors, Wakefeta Religion and no other income generating activities were
statistically significantly associated with home delivery when ANC was not included in the model. When
ANC was included, none of the variables retained statistical significance. The odds of attending ANC by
those who delivered at home was 19% of that did not attend ANC. Regarding religion and other income
generation activities, the association with home delivery can be explained in a similar manner as that of
ANC attendance, but in the reverse order or direction. That is belonging to Wakefeta Religion and not
having other income generating activities do not enable visiting health facilities for services.
35
Birth registration within 14 days is expected to enhance polio birth dose vaccination, as it helps identify
newborns and/ or vaccinate them with birth doses as they are identified. As in the case of deliveries
none of the other factors had significant association with birth registration when ANC is included in the
model showing the lofty importance of ANC attendance.
Polio was the most frequently known vaccine preventable disease, but about half of the respondents did
not mention it. Second comes measles but only 44% of the respondents had mentioned it. Moreover, only
29% of the women said that vaccination starts at birth. These indicate the gap that prevails in
comprehensive and necessary knowledge about vaccine preventable diseases and vaccination schedule.
Vaccination status of children is difficult to measure because of difficulties in acquiring information about
whether the child had received each antigen included in the national immunization program.
Community based data collection employs two approaches; interviewing mothers or care takers about
whether the child has gotten the required antigen for his/her age and referring to the vaccination card of
the child. The methods have their advantages and disadvantages. Referring to the card is a more objective
way of assessing immunization status because interviewing the mother is dependent on her memory and
understanding the schedule and the routes of vaccine administration. On the other hand, vaccination
cards are often lost. In this study we employed both methods and a combination of the methods (and/or)
so that data would be analyzed using any of the sources. Use of interview had a higher coverage rate
compared to cards. A combination of the two yields the highest coverage, but perhaps exaggerates it.
Accordingly, this study revealed that BCG coverage was 53.7%, 40.8%, 57.2% by history, by card and by
history or card respectively. Polio 0 coverage was 29.7%, 19.7%, and 32.7% by history, by card and by
history or card respectively. Penta 3 coverage was 33.8%, 27.3% and 39% by history, by card and by
history or card respectively. This finding is similar to a study in disadvantaged (slum) areas in Nairobi,
Kenya that reported a much lower coverage of Polio 0 was reported compared to other antigens (12 )
According to the 2010/2011 FMOH report, Penta 3 coverage in Afar was 37.4%, Oromiya 86%,
Somali 34%, Ben gumuz, 98%, Gambella 70.4%, which generally indicates a higher coverage than the
study districts. At this juncture it is worth noting that the FMOH annual report does not include polio 0
coverage reports, which is an impediment in follow up of progress in Polio 0 status and trend. On the
other hand the 2011 Ethiopian Demographic Health Survey indicates generally lower coverage rates.
Coverage rates for Polio 0 by mother’s report or card for Afar, Somali, Benishagul Gumuz, Gambella
and Ormiya regions were 10.6%, 18.9%, 36.4%, and 15.5 polio respectively. Penta 3/DPT3 coverage
reported by the Demographic and Health survey were 10.3%, 25.3%, 27.8%, and 26.8% for Afar, Somali,
Benshangul-Gumuz, Gambella and Oromiya regions respectively. While differences in coverage rates
between FMOH reports that are based on health service data and the community based demographic and
health survey are not unexpected, the observed wide gap cannot be explained and might warrant further
36
studies. Nonetheless, the wide gap renders it difficult understanding the status of immunization in the
study districts (woredas) compared to national or local coverage. As also mentioned above, it should be
noted that the study districts (woredas) may not be representative of the administrative region in which
they are located because of selection criteria, interventions and other reasons and this also applies to
vaccination coverage.
Religion and attending ANC, non-home delivery (health facility delivery) birth registration within 14
days of delivery were statistically significantly associated with polio birth dose vaccination. This finding,
although may be already expected, has an important implications of the chain of actions that need to be
undertaken for improving polio birth dose coverage.
The reasons that mothers gave for no polio birth dose of the newborn indicate lack of knowledge such as
not knowing the importance of polio vaccination, not knowing when polio vaccination starts and does
not know the site of vaccination. It also indicates issues related to inaccessibility of services as shown by
responses such as unavailability of services during their visit, distance from the health facility
inconvenient service time and poor health workers’ approach. Other reasons although mentioned less
frequently refer to societal or personal situation of the mother that may not be directly related to the
provision of services. These include mother was too busy, delivery took place not in the normal (current
residence area), fear of side effects, and mother can’t go out of home soon after delivery. The implication
of these findings is that interventions need to be undertaken from different perspectives; informing and
educating women and improving accessibility and quality of services. A pilot birth dose project in India
showed some similarity in reasons for refusal of polio birth dose such as fear of side effects (sterility) and
misunderstanding about when and how many doses of polio vaccine are needed (13).
Similar conclusions can be made about the need for educating women on polio transmission and
prevention because of the following and related findings. Forty to seventy percent of the respondents
answered that they did not know how polio is transmitted and prevented. Not a small number of
respondents also gave incorrect answers about polio transmission such as sneezing and cough and polio
vaccination starts at one month of age.
Forty one (58.6%) of the HEWs were female and 29 (41.4%) were male. This is different from agrarian
parts of the country where all HEWs are female. Having female HEWs, among other things was justified
by issues related to acceptance, since many of their activities deal with women and families. The
implications of having male HEWs in the study areas from performance, acceptance, and feasibility
points of view need to be explored further.
37
Interviews with HEWs showed that less than one third (31.4%) of them were trained in Newborn
Tracking and 5(7.1 %) in social mobilization indicating the need for training in order to improve polio
birth dose coverage. Sixty (85.7%) HEWs reported to have received supervision during the previous of
six months. However, only 39(63.9%) got feedback, indicating another area that needs to be strengthened.
Similar experiences were reported by the CVSFPs.
The role of TBAs, trained or untrained, in conducting deliveries in developing countries has been
controversial and it seems that nowadays there is a consensus that they should not handle child
deliveries on their own. That also appears to be the stand of the Ministry of Health of Ethiopia, where
institutional delivery is being highly promoted recently. However, the number of reported deliveries
conducted by the TBAs ranged from2 to 40 during the previous 3-6 months, corroborating with the
finding from the women’s interview that a large majority of deliveries are conducted by TBAs. Thus, it
becomes impossible to exclude the participation of TBAs if improving mother and newborn health is to
take place in such settings. However, only4 of the 47 TBAs knew that the first dose of Polio (birth dose)
is given at birth or within 14 days after birth indicating a huge missed opportunity for informing and
motivating mothers for polio birth dose immunization.
Concerning polio birth dose vaccination, TBAs’ suggestions coincide with those of HEWS and CVSFPs.
The TBAs have also recommended closer ties between TBAs and health workers, and informing HEWs or
other health workers about deliveries that have high implications for improving polio birth dose
coverage. However, a constraint in implementing the latter may be that TBAs may not be officially
allowed to conduct deliveries and hence would be unable to report about the deliveries. Thus the
approach to be used and the roles of TBAs should be defined in the specific context.
None of the health center EPI coordinators knew Polio 0 coverage of their catchment area. They also did
not know ANC and Institutional delivery coverage. Only 2 woreda EPI coordinators could provide
figures on Polio 0 coverage. This indicates the large gap that needs to be filled in improving awareness of
the program owners (EPI coordinators) on the importance of polio birth dose, its documentation and
interpretation, by having a system of documentation of polio birth dose, training and supportive
supervision. Concerning improving Polio birth dose coverage, they shared the suggestions mentioned by
other respondents viz increase institutional delivery, educate the community on Polio birth dose, house
to house search and vaccination by HEWs, provision of polio birth doses at outreach sites, avail
refrigerators, supplies and vaccines at all health facilities, and networking with TBAs and volunteers.
This implies that there is a consensus on what should be done to improve polio birth dose coverage.
Several participants in the FGDs mentioned that polio causes paralysis of legs and inability to walk. On
the other hand there are indications about possible misconceptions on the causes and transmission of
38
polio virus that need to be curved by IEC. These misconceptions include leg swelling and fainting as
symptoms of polio, polio caused by curse, polio transmitted airborne from patients, body contact with a
polio patient and rumors such as vaccination may cause sterility. Almost all participants agreed that there
are no reasons related to religious beliefs that discourage immunization in the study areas. This finding
has a very important implication for increasing polio birth dose vaccination, since religious institutions
and leaders can inform and motivate mothers and families for vaccination if they are informed and their
commitment is secured. On the other hand, it was found that among women who delivered babies
during the previous one year followers of the Wakefeta religion had a lower rate of immunization
compared to other religions. The reasons for such a low rate when all religions were said to encourage
immunization should be explored and special attention needs be given to the followers of the Wakefeta
religion.
Once again the measures cited by the FGD participants to improve polio birth dose were similar to those
of other study populations. More or less new additions that have important implication are involving
elders and religious leaders in promoting newborn vaccination and creating mechanisms for kebele
administration to take responsibility for birth dose vaccination, using all available venues such as
mosques, churches and market places to inform about Polio birth dose.
Strengths and limitations of the study
This study used several study populations and a combination of qualitative and quantitative methods
and presented a comprehensive answer for the study questions from different perspectives.
However, it cannot claim to be representative of all pastoralist and semi-pastoralist areas of the country,
as the different areas may have unique characteristics related to health seeking and related issues.
Similarly, each study district may have certain unique features that may not have been well covered in
the study. Moreover, some extremely hard to reach areas and those with security problems during the
time of data collection were excluded from the study and hence the findings may not reflect the realities
in these areas. In addition, as mentioned above, certain variables, especially those related to assessment of
economic status were not included in the results of the study.
39
CONCLUSION
This study took place in a sample of CORE Group Polio Project pastoralist and semi-pastoralist areas
characterized by high illiteracy rate, almost universal marriage where monogamy and polygamy are
practiced, gender mix of HEWs , highly influential community and religious leaders, low ANC and
institutional delivery , low birth registration and low polio birth dose rates.
Religion and attending ANC, non-home delivery (health facility delivery) birth registration within 14
days of delivery were statistically significantly associated with polio birth dose vaccination. This finding
has an important implication of the chain of actions that need to be undertaken for improving polio birth
dose coverage.
A further analysis on determinants of ANC showed the following. Religion, having live stock, and
income generating activities were significantly associated with ANC attendance. Nonetheless, being able
to read and write easily did not maintain significant association with ANC attendance when adjusted for
other factors. After adjusting for other factors, Wakefeta Religion and no other income generating
activities were statistically significantly associated with home delivery when ANC was not included in
the model. When ANC was included, none of the variables retained statistical significance. Birth
registration within 14 days is expected to enhance polio birth dose vaccination, as it helps identify
newborns and/ or vaccinate them with birth doses as they are identified. Similar, but inverse associations
as that of home delivery were found. As in the case of deliveries none of the other factors had statistically
significant association with birth registration when ANC is included in the model showing the lofty
importance of ANC attendance. Thus improving ANC attendance and by extension; identification,
registration and follow up pregnant women should be given high priority to improve polio birth dose
coverage.
Based on the responses of women who delivered in the previous one year, the factors that lead to low
birth dose coverage can be summarized as lack of knowledge on prevention, transmission of polio and
schedule for vaccination, inaccessibility and poor quality of services and societal or individual related
factors. The implication of having such diverse findings is that interventions need to be undertaken from
different perspectives; health services and community level.
Interviews with HEWs and CVSFPs showed that less than one third of them were trained in Newborn
Tracking and social mobilization and supervision was not always followed by feedback and follow up
indicating the need for training and strengthening supportive supervision.
40
The majority of the deliveries were conducted by TBAs as shown by the women who delivered in the
previous one year and TBAs’ interviews. Thus, although the role of TBAs has been controversial
nowadays, it becomes impossible to ignore the participation of TBAs in improving mother and newborn
health in the study area.
None of the health center EPI coordinators knew Polio 0 coverage of their catchment area. They also did
not know ANC and institutional delivery coverage. Only 2 woreda EPI coordinators could provide
figures on Polio 0 coverage. It is thus necessary to train these “program owners” ( Woreda and Health
center EPI coordinators) on the importance of polio birth dose documentation and interpretation, and
have an improved system of documentation of polio birth dose and supportive supervision.
Almost all FGD participants stated all religions encourage immunization in the study areas. This
opportunity can and must be used to improve polio birth dose vaccination because religious leaders and
institutions are highly respected and their advice and instructions firmly followed in the study areas.
A strikingly similar set of suggestions was made by the different study population groups related to
improving awareness, service provision and participation of stakeholders in improving polio birth dose
coverage. These have been incorporated in the recommendations section that follows.
RECOMMENDATIONS
1. Improving awareness of women, families and communities through targeted IEC
interventions
Raise awareness of women, family and community leaders on causes, transmission , prevention
of polio and vaccination schedule emphasizing on polio birth dose
Prepare messages and materials based on identified gaps, misconceptions or incorrect responses
on the issues mentioned above ie causes, transmission , prevention of polio and vaccination
schedule emphasizing on polio birth dose
Use appropriate mechanisms to reach women, families and communities with messages. This
may include house to house HEW and CV visits, using TBAs contacts with pregnant and
delivering women, community meetings , local media, religious and other institutions .
2. Training and strengthening of supportive supervision
Train or retrain HEWs, CVs, TBAs on newborn tracking and social mobilization
Develop mechanisms for effective implementation of planning , conducting and monitoring of
supportive supervision and feedback of HEWs and CVs within the existing health and
administration systems
41
Train Woreda and Health Center EPI coordinators with particular emphasis on the importance of
polio birth dose documentation, interpretation and utilization of results and developing such a
system at different levels, as appropriate and feasible.
3. Develop and strengthen mechanisms for identification and follow up of pregnant women, use
of ANC and institutional delivery, birth registration and subsequent polio birth dose
vaccination
House to house identification of pregnant women by HEWs, CVs
Link with TBAs for identification and reporting of pregnant women and women who have just
delivered
Involve religious leaders and community elders and their institutions on each of the steps
mentioned above.
Involve kebele or other formal leaders to take responsibility on improving polio birth dose
coverage
4. Develop and strengthen strategies to improve accessibility and quality of maternal and newborn
health services
Improve provision of essential equipment and supplies, with particular emphasis to health
posts.
Expansion of immunization service delivery such as outreaches.
Training on management of resources as appropriate
Support advocacy activities that target decision makers
42
REFERENCES
1. Central Statistical Agency (Ethiopia) and ORC Macro. Ethiopia Demographic and Health Survey
20010/11, Addis Ababa, Ethiopia and Calverton, Maryland, USA: Central Statistical Agency and
ORC Macro, 2012
2. Central Statistical Agency (Ethiopia) and ORC Macro. Ethiopia Demographic and Health Survey
2005, Addis Ababa, Ethiopia and Calverton, Maryland, USA: Central Statistical Agency and ORC
Macro, 2006 .
3. Federal Democratic Republic of Ethiopia. Ministry of Health. HSDP IV. Annual Performance
Report EFY 2003 ( 2010/2011).
4. CORE Group Ethiopia. Survey of Immunization of children and women in CORE Group
partners’ implementation areas. March 2010.
5. CORE Group Polio Project in Angola, Ethiopia, and Uttar Pradesh, India: Midterm Evaluation.
June 30. 2011.
6. CORE Group Polio Project Midterm Evaluation in Ethiopia. December 2010.
7. CORE Group Ethiopia. AFP case detection and status of surveillance in pastoralist and semi-
pastoralist communities of CORE Group Polio Project implementation districts (woredas) in
Ethiopia. June 2012.
8. CORE Group Ethiopia. Cross Border Transmission of Wild Polio Virus ( WPV) and
Immunization Service Delivery in CGPP Project Implementation International Border Areas in
Ethiopia. June 2012.
9. Lemeshow S, Taber S Lot quality assurance sampling: single- and double-sampling plans. World
Health Stat Q. 1991;44(3):115-32.
10. Singh J, Jain DC, Sharma RS, Verghese T. Evaluation of immunization coverage by lot quality
assurance sampling compared with 30-cluster sampling in a primary health centre in India. Bull
World Health Organ. 1996;74(3):269-74.
11. Federal Democratic Republic of Ethiopia. Ministry of Health. Health and Health Related
Indicators. For 2003 Ethiopian Fiscal Year (2010/2011 GC). 2012.
12. Mutua M, Kimani-Murage E, Ettarh R. Childhood vaccination in informal urban settlements in
Nairobi, Kenya: Who gets vaccinated? BMC Public Health 2011, 11:6.
13. Rainey J, Bhatnagar P, Estivariz CF et al.. Providing monovalent oral polio vaccine type 1 to
newborns: findings from a pilot birth-dose project in Moradabad district, India. Bull World
Health Organ. 2009; 87:955-959. doi: 10.2471/BLT.08.061556